Issue Date March 2, 2011 Audit Report Number 2011-BO-1005 TO: Vicki Bott, Deputy Assistant Secretary for Single Family Housing, HU FROM: John A. Dvorak, Regional Inspector General for Audit, Boston Region, 1AGA SUBJECT: Semper Home Loans, Inc., Providence, RI, Needs To Improve Its Quality Control Process for Loan Origination and Updating of Mortgage Records HIGHLIGHTS What We Audited and Why We audited Semper Home Loans, Inc. (Semper), a Federal Housing Administration (FHA) lender approved to underwrite and close mortgage loans without prior FHA review or approval.1 We selected Semper because its early payment default rate was higher than the default rate in the local area in which it does business.2 Our audit objectives were to determine (1) whether Semper acted in a prudent manner and complied with U.S. Department of Housing and Urban Development (HUD) regulations, procedures, and instructions for the origination, underwriting, and closing of the FHA-insured single-family loans selected for a detailed review and (2) whether Semper’s quality control plan, as implemented, fulfilled HUD’s requirements. 1 On April 14,2010, the lender became a full underwriting lender and changed its name from Semper Financial Mortgage Corporation to Semper Home Loans, Inc. Prior to that, the lender operated as a loan correspondent beginning from March 22, 2007. 2 For insured single-family loans originated between January 1, 2008, and December 31, 2009 What We Found Semper, as a former loan correspondent and current direct endorsement lender, generally met HUD requirements for the origination of FHA-insured single- family loans. However, we identified several underwriting deficiencies that negatively affected the insurability of two loans for which Semper acted as the loan correspondent and for which the underwriting was performed by one of Semper’s sponsors (Fairfield Financial Mortgage Group, Inc.). These deficiencies occurred because the underwriter did not act in a prudent manner when approving the two loans for FHA-insurance. These deficiencies are not directly attributable to Semper but need to be addressed. The underwriting deficiencies for both loans consisted of qualifying ratios that exceeded HUD’s benchmarks without significant compensating factors. In addition, Semper did not fully implement its quality control plan, although the plan met all of HUD’s requirements. It failed to perform sufficient reviews, which prevented it from ensuring the accuracy, validity, and completeness of its loan origination operations. This deficiency occurred because Semper’s management was not fully aware of HUD’s requirements for following its quality control plan. As a result, Semper may not have identified and corrected potential deficiencies in a timely manner, resulting in an unnecessary risk to the FHA insurance fund. Semper was also incorrectly listed as the holding lender for 34 active loans and the servicing lender for 11 active loans. These errors occurred because Semper was not aware of HUD requirements for mortgage record changes after it sold loans to investing lenders. Inaccurate or untimely reporting of mortgage record changes directly affects the payment of claims for insurance benefits. HUD will not pay a claim for insurance benefits for which the information on the claim and HUD’s FHA insurance system does not agree. What We Recommend We recommend that HUD’s Deputy Assistant Secretary for Single Family Housing require the sponsor(s) of the respective loans to (1) reimburse the FHA insurance fund $169,000 in actual losses on one loan and (2) indemnify HUD for a potential loss of $179,400 that may be incurred related to one loan that did not meet FHA insurance requirements. We also recommend that HUD’s Deputy Assistant Secretary for Single Family Housing direct Semper Home Loans to (1) implement its quality control plan as required and follow up with the lender in 9 months to ensure its compliance and (2) update its remaining mortgage records in HUD’s system to reflect the 2 appropriate mortgage holder and implement procedures to ensure the timely submission of mortgage record changes for future loans sold to investing lenders. For each recommendation without a management decision, please respond and provide status reports in accordance with HUD Handbook 2000.06, REV-3. Please furnish us copies of any correspondence or directives issued because of the audit. Auditee’s Response We provided Semper officials a draft report on February 14, 2011, and requested a response by February 25, 2011. We discussed the draft report at an exit conference on February 17, 2011, and received Semper’s written comments on February 21, 2011. Semper agreed with the results of the audit. HUD also provided comments. The complete text of the auditee’s and HUD’s response, along with our evaluation of the responses, can be found in appendix B and C of this report. 3 TABLE OF CONTENTS Background and Objectives 5 Results of Audit Finding 1: A Sponsor of Semper Did Not Underwrite Two Loans in Accordance 7 With HUD Requirements Finding 2: Semper’s Implementation of Its Quality Control Plan Was Deficient 11 Finding 3: Mortgage Records Were Not Accurate in HUD Systems 15 Scope and Methodology 17 Internal Controls 19 Appendixes A. Schedule of Questioned Costs and Funds To Be Put to Better Use 21 B. Auditee Comments and OIG’s Evaluation 22 C. HUD Comments and OIG’s Evaluation 25 D. Loan Details 27 4 BACKGROUND AND OBJECTIVES The National Housing Act, as amended, established the Federal Housing Administration (FHA), an organizational unit within the U.S. Department of Housing and Urban Development (HUD). FHA provides insurance to protect lenders against losses on mortgages financing homes. The basic single-family mortgage insurance program is authorized under Title II, Section 203(b), of the National Housing Act and is governed by regulations in 24 CFR (Code of Federal Regulations) Part 203. The single-family programs are generally limited to dwellings with one- to four-family units. HUD handbooks and mortgagee letters provide detailed processing instructions and advise the mortgage industry of major changes to FHA programs and procedures. Semper Home Loans, Inc. (Semper) is a nonsupervised3 mortgage company. Semper became an approved Title II loan correspondent4 authorized to originate FHA loans on March 22, 2007. On April 14, 2010, the lender became a full underwriting lender and changed its name from Semper Financial Mortgage Corporation to Semper Home Loans, Inc. and also moved to a new address, 225 DuPont Drive, Providence, RI. During the audit, in addition to its main location in Providence, RI, Semper operated an active branch in Charlotte, North Carolina. Subsequently, as of February 2011, Semper added another branch based in Morristown, NJ. The lender does not sponsor any loan correspondents. It is an authorized agent for two lenders and authorized principal for eight lenders. We identified Semper as a lender for review based on a risk assessment of mortgage lenders in the New England region. We identified Semper as having a higher than average FHA-insured mortgage default rate when compared to other FHA lenders. The lender originated and underwrote 453 loans during our review period (January 1, 2008 through December 31, 2009) with a total original mortgage amount of more than $100 million. It originated at least one FHA loan in eight different States during this period, with primary originations occurring in Massachusetts, New York, and Rhode Island. Thirty-Seven of the loans (or 8.17 percent) defaulted within the first 2 years of origination. There were no claim terminations.5 When comparing loans underwritten by the lender to the rest of the lenders in each State, the lender had a total early payment default percentage that was much higher than average (compare ratio),6 especially for those loans originated in the New England region (see table below). 3 A nonsupervised lender is a financial institution that has as its principal activity the lending or investment of funds in real estate mortgages. A nonsupervised lender can originate, purchase, hold, service FHA insured loans, and submit applications for insurance. 4 A loan correspondent is a mortgagee, which has as its principal activity the origination of FHA insured loans for the sale or transfer to its sponsor(s) for underwriting. A loan correspondent can originate and sell (to its sponsors) FHA insured loans, and submit applications for insurance. 5 Claim terminations occur when the lender submits a claim to obtain insurance benefits from HUD resulting in termination of the FHA insurance. There are a total of 11 claim types. 6 The percentage of originations that were seriously delinquent or were claim terminated divided by the percentage of originations that were seriously delinquent or were claim terminated for the selected geographic area. Compare ratio is the value that reveals the largest discrepancies between the subjects’ seriously delinquent and claim percentage and the seriously delinquent and claim percentage to which it is being compared. 5 State State State Total # of def % of def total State # of def % of def by Compare Total defaults % def by 2 yr by 2 yr State defaults % def by 2 yr 2 yr to State ratio orig. by 2 yr by 2 yr to claim to claim total orig by 2 yr by 2 yr to claim claim Rhode Island 238% 74 9 12.16 0 0.00 10,552 538 5.10 28 5.20 Florida 214% 9 2 22.22 0 0.00 146,205 15,188 10.39 187 1.23 New York 198% 86 11 12.79 0 0.00 90,184 5,818 6.45 29 0.50 Connecticut 133% 52 4 7.69 0 0.00 35,641 2,065 5.79 45 2.18 Massachusetts 123% 91 5 5.49 0 0.00 46,024 2,051 4.46 39 1.90 Virginia 75% 26 1 3.85 0 0.00 107,453 5,495 5.11 360 6.55 Georgia 62% 52 3 5.77 0 0.00 135,686 12,546 9.25 674 5.37 North Carolina 47% 63 2 3.17 0 0.00 97,374 6,584 6.76 255 3.87 *Source: HUD’s Neighborhood Watch/Early Warning System (January 1, 2008 through December 31, 2009) The lender’s loan volume increased significantly from 2007 to 2009. The majority of loans originated by the lender were refinance transactions (more than 93 percent - 452/485). The audit objectives were to determine whether (1) Semper acted in a prudent manner and complied with HUD regulations, procedures, and instructions for the origination, underwriting, and closing of the FHA-insured single-family loans selected for a detailed review and (2) its quality control plan, as implemented, fulfilled HUD’s requirements. 6 RESULTS OF AUDIT Finding 1: A Sponsor of Semper Did Not Underwrite Two Loans in Accordance With HUD Requirements Of the twenty-two loans selected for review, we identified two loans that were not underwritten in accordance with HUD requirements by a sponsor of Semper. Specifically, the two loans exhibited underwriting deficiencies significant enough to warrant indemnification. In both instances, the underwriter of the sponsor did not provide significant compensating factors to mitigate the high debt-to-income ratios as required. When loans exceed the standard debt-to- income ratio, FHA regulations require the lender to obtain and document significant compensating factors. These deficiencies occurred because the underwriter did not act in a prudent manner when approving the two loans for FHA-insurance. Although proper HUD underwriting guidelines were not followed for these two loans, there was no indication of a pattern of noncompliance. However, the FHA insurance fund incurred losses totaling $169,000 and is at increased risk for an additional loss of $179,400. Two Loans Had Significant Underwriting Deficiencies We identified and conducted a detailed review of twenty-two FHA-insured loans. A sponsor of Semper, however, did not underwrite two loans in accordance with HUD requirements, and the loans had underwriting deficiencies that warranted indemnification. Semper, as the loan correspondent, simply originated the two loans. Specifically, the sponsoring lender, Fairfield Financial Mortgage Group, Inc., did not document compensating factors for each loan as required. When the standard debt-to-income ratios exceed HUD guidelines, FHA regulations require the lender to obtain and document significant compensating factors. HUD established benchmark guidelines setting the two qualifying ratios, the housing expense ratio and the total expense ratio, at 31 and 43 percent, respectively. Ratios exceeding these thresholds may be acceptable only if significant compensating factors are documented and recorded.7 FHA Case No. 451-0946313 For FHA case no. 451-0946313, the sponsor correctly calculated both the qualifying housing expense ratio and total expense ratio to be 46.4 percent, 7 HUD Handbook 4155.1, section 4.F and Mortgagee Letter 2005-16. 7 significantly exceeding the benchmarks established by HUD. The relationship of the mortgage payment to income is considered acceptable if the total mortgage payment does not exceed 31 percent of the gross effective income. A ratio exceeding 31 percent may be acceptable only if significant compensating factors are documented and recorded on Form HUD-92900-LT, FHA Loan Underwriting and Transmittal Summary8. In addition, the relationship of total obligations to income is considered acceptable if the total mortgage payment and all recurring charges do not exceed 43 percent of the gross effective income. A ratio exceeding 43 percent may be acceptable only if significant compensating factors are documented and recorded on Form HUD-92900-LT. Compensating factors that are used to justify approval of mortgage loans with ratios that exceed benchmark guidelines must be recorded on the underwriter comments section of Form HUD-92900-LT. Any compensating factor used to justify mortgage approval must also be supported by documentation.9 No compensating factors were documented. No other deficiencies were noted. This loan was a conventional refinance with an initial mortgage of $308,560, and a closing date of June 30, 2008. As of January 26, 2011, the latest data pulled from HUD’s Neighborhood Watch/Early Warning System showed that the borrower made just four payments before first becoming 90-days delinquent, and was thirteen months delinquent with an unpaid principal balance of $299,000. The cause of delinquency was listed as “curtailment of borrower income” and a pre-foreclosure sale was held. A pre-foreclosure sale allows the borrower to sell the home at fair market value, which may be less than the amount owed to the lender, and HUD then reimburses the lender the difference between the sales proceeds and the outstanding mortgage indebtedness. No claims have been paid by HUD. See appendix C for loan details. FHA Case No. 451-0941164 For FHA case no. 451-0941164, the lender incorrectly calculated the qualifying housing expense ratio and total expense ratio at 37.8 and 49.3 percent, exceeding the benchmarks established by HUD. It is unclear from the files reviewed, what data the lender used to calculate the ratios. Therefore, using the information contained within the loan file, we recalculated the ratios and determined that the qualifying housing expense ratio was 43.3 percent and the total expense ratio was 54.8 percent, both significantly exceeding the benchmarks established by HUD as outlined in the previous case. Any compensating factor used to justify mortgage approval must also be supported by documentation.10 No compensating factors were documented. No other deficiencies were noted. 8 Form HUD-92900-LT is a form used by the underwriter to record the results of the credit analysis of an approved borrower. Any modifications of the mortgage amount or approval conditions are reflected in the “Underwriter Comments” section of the form. By signing and dating this form (when required),underwriters are providing their final decision to approve the loan application for FHA mortgage insurance. 9 HUD Handbook 4155.1, section 4.F and Mortgagee Letter 2005-16 10 HUD Handbook 4155.1, section 4.F and Mortgagee Letter 2005-16 8 This loan was a conventional refinance with an initial mortgage of $315,056, and a closing date of July 16, 2008. As of January 26, 2011, the latest data pulled from HUD’s Neighborhood Watch/Early Warning System showed that the borrower made just five payments before first becoming 90-days delinquent, and was ten months delinquent with an unpaid principal balance of $309,686. The cause of delinquency was listed as “curtailment of borrower income” and a pre- foreclosure sale has been completed, resulting in HUD paying a claim of $169,000 on January 29, 2010. The insurance has been terminated and HUD will not incur any additional losses. See appendix C for loan details. The Sponsor Did Not Act in a Prudent Manner The sponsor did not act in a prudent manner when it approved the two loans for FHA-insurance. It should have been more prudent when evaluating the documentation provided and reviewing the qualifying ratios, considering that they were significantly higher than the benchmarks set by HUD. Conclusion A sponsor of Semper did not underwrite two loans in accordance with HUD requirements. Specifically, the two loans exhibited underwriting deficiencies significant enough to warrant indemnification. When loans exceed the standard debt-to-income ratio, FHA regulations require the lender to obtain and document significant compensating factors. These deficiencies occurred because the underwriter did not act in a prudent manner when approving the two loans for FHA-insurance. As a result, the FHA insurance fund incurred losses totaling $169,000 and is at increased risk for an additional loss of $179,400. Recommendations We recommend that HUD’s Deputy Assistant Secretary for Single Family Housing require the sponsor (Fairfield Financial Mortgage Group, Inc.) of Semper to 1A. Indemnify HUD for one insured loan (FHA case no. 451-0946313) with an unpaid principal balance of $299,00011, thereby putting an estimated $179,400 11 Data obtained on January 26, 2011, from HUD’s Neighborhood Watch/Early Warning System. 9 to better use based on the FHA insurance fund average loss rate of 60 percent of the unpaid principal balance12. 1B. Reimburse the FHA insurance fund $169,000 for losses incurred on FHA case no. 451-0941164. 12 HUD’s estimated loss is computed using FHA’s FY 2009 Actuarial Review of the Mutual Mortgage Insurance Fund. The average loss experienced is about 60% of the unpaid principal balance upon the sale of a mortgaged property. 10 RESULTS OF AUDIT Finding 2: Semper’s Implementation of Its Quality Control Plan Was Deficient Semper did not routinely conduct timely quality control reviews required by its plan and HUD’s requirements. Although Semper’s quality control plan included all of the necessary HUD requirements, Semper did not ensure that all quality control reviews were performed on a regular, timely basis. This deficiency occurred because Semper’s management was not fully aware of HUD’s requirements for following its quality control plan. The failure to perform sufficient reviews prevented Semper from ensuring the accuracy, validity, and completeness of its loan origination operations. As a result, it may not have identified and corrected potential deficiencies in a timely manner, resulting in an unnecessary risk to the FHA insurance fund. Quality Control Implementation Had Deficiencies Semper’s implementation of its quality control plan had several deficiencies. Semper did not Perform quality control reviews within 90 days of loan closing for randomly selected loans, Perform quality control reviews for all loans going into default within 90 days of loan closing, Perform quality control reviews of a sample of rejected loans, Always obtain complete credit documentation reverification, or Require field reviews of appraisals by licensed appraisers. Quality Control Reviews Are Required Within 90 Days of Loan Closing HUD requires that lenders perform quality control reviews within 90 days of closing on at least a quarterly basis.13 This requirement is intended to ensure that problems left undetected before closing are identified as early after closing as possible. Of the 33 quality control reviews, 24 were not performed within the required 90 days from the end of the month of closing. According to the audit 13 HUD HB 4060.1 REV-2, Chapter 7, 7-3D 11 response for the second, third, and fourth quarters of 2009, the third-party firm performing the quality control reviews, suddenly and without notice, dissolved and referred Semper to another firm. Semper searched for other firms to conduct its quality control reviews but in the end, settled on the referred firm because it promised to honor the pricing agreement and already had the second quarter of 2009 loan files in hand. Therefore, Semper fell several months behind the anticipated schedule. However, in 2010 the quality control reviews were also not performed in a timely manner. The first quarter reviews were performed on May 26, 2010. Two of the three quality control reviews performed for the first quarter of 2010 were also not performed within 90 days from the end of the month of closing (these two are included in the total 24 discussed above). The second quarter reviews only included April and May 2010 and were performed on September 30, 2010, which was well over 90 days after the end of the month of closing. Semper started underwriting its own loans in May/June 2010 so it went to monthly reviews as of June 2010. As of October 4, 2010, it had not received the June 2010 reviews from its quality control contractor. In addition to the loans selected for routine quality control reviews, lenders must review all loans going into default within the first six payments. Early payment defaults are loans that become 60 days past due.14 Semper officials were not aware of this requirement for loan correspondents. Semper’s executive vice president stated that its sponsors had to review these loans but did not believe that Semper also had to do so. According to HUD regulations, loan correspondents may arrange with their sponsor(s) to perform quality control provided (1) the arrangement with the sponsor(s) is detailed in writing, (2) the aggregate number and scope of reviews meet FHA requirements, (3) loans are reviewed within 90 days of closing, (4) findings are clear as to source and cause, and (5) results are available in a timely manner to both lenders and HUD.15 Semper did not have an agreement with any of its sponsors to perform its quality control reviews. According to Semper’s vice president, early payment default loans were part of its quality control plan and would be reviewed now that Semper is a lender. Quality Control Reviews Are Required for Rejected Loans HUD also requires that lenders perform quality control reviews of rejected loans. A minimum of 10 percent or a random sample that provide a 95 percent 14 HUD HB 4060.1 REV-2, Chapter 7, 7-6D 15 HUD HB 4060.1 REV-2, Chapter 7, 7-3, H.2 12 confidence level with a 2 percent precision of reject loans must be reviewed, concentrating on the following areas:16 Ensuring that the reasons given for rejection were valid; Ensuring that each rejection has the concurrence of an officer or senior staff person of the company or a committee chaired by a senior staff person or officer; Ensuring that the requirements of the Equal Credit Opportunity Act are met and documented in each file; Ensuring that no civil rights violations are committed in rejection of applications; and When possible discrimination is noted, taking immediate corrective action. Semper’s management was not aware that rejected loans needed to be reviewed as part of the quality control process. This requirement is in Semper’s quality control plan and needs to be implemented to ensure that Semper complies with the fair lending laws. Sufficiency and Reverification Were Not Clearly Documented HUD requires that documents contained in the loan file be checked for sufficiency and subjected to written reverification. Examples of items that must be reverified include but are not limited to the borrower’s employment or other income, deposits, gift letters, alternate credit sources, and other sources of funds. If the written reverification is not returned to the lender, a documented attempt must be made to conduct a telephone reverification.17 Semper did not clearly document that these requirements were followed. Its quality control firm sent a letter to the employer reported at the time of the closing and asked it whether it had completed the attached verification of employment and to confirm the accuracy of the information. Some of these forms were not completed. For others, it appeared as though the same person did not complete the form. 16 HUD HB 4060.1 REV-2, Chapter 7, 7-8, A.1 17 HUD HB 4060.1 REV-2, Chapter 7, 7-6, E.2 13 Field Reviews by Licensed Appraisers Were Not Conducted We also did not find any indication of field reviews by licensed appraisers before Semper’s becoming a direct endorsement lender. A desk review of the property appraisal must be performed on all loans chosen for a quality control review except streamline refinances and HUD real estate owned sales. Lenders are also expected to perform field reviews of 10 percent of the loans selected during the sampling process.18 Semper had started including a review of the appraisals since it is now a direct endorsement lender. Conclusion Semper did not fully implement a quality control plan, although it met all of HUD’s requirements. This deficiency occurred because Semper’s management was not fully aware of all of HUD’s quality control plan requirements. Semper’s current quality control plan included all of the necessary requirements; however, Semper needs to ensure that all of the requirements are met by ensuring that quality control reviews are routinely conducted on time. The failure to perform sufficient reviews prevented Semper from ensuring the accuracy, validity, and completeness of its loan origination operations. As a result, Semper may not have identified and corrected potential deficiencies in a timely manner, resulting in an unnecessary risk to the FHA insurance fund. Recommendations We recommend that HUD’s Deputy Assistant Secretary for Single Family Housing require Semper to 2A. Implement its quality control plan as required and follow up with the lender in 9 months to ensure its compliance. 18 HUD HB 4060.1 REV-2, Chapter 7, 7-6, E.3 14 RESULTS OF AUDIT Finding 3: Mortgage Records Were Not Accurate in HUD Systems Semper was incorrectly listed as the holding lender for 34 active loans and the servicing lender for 11 active loans. This condition occurred because Semper was not aware of HUD requirements regarding mortgage record changes. Inaccurate or untimely reporting of mortgage record changes directly affects the payment of claims for insurance benefits. HUD will not pay a claim for insurance benefits for which the information on the claim and HUD’s FHA insurance system does not agree. Mortgage Records for Semper Were Not Accurate As of October 31, 2010, Semper was still listed as the holding lender for 34 active loans and the servicing lender for 11 active loans, most of which were more than 90 days past endorsement. Semper sells all loans that it originates, including the servicing rights, at closing to investing lenders. Originating lenders initially process the loan application. Holding lenders hold title to the mortgage note. Servicing lenders maintain the servicing rights to the loan as they relate to FHA-insured mortgages, including the collection of loan payments, servicing delinquent accounts, foreclosure processing, mortgage insurance premium billing, escrow administration, and general maintenance of records . In November 2003, recognizing the new technology under which the mortgage industry and HUD operates the single-family insurance programs, HUD eliminated the paper mortgage insurance certificate in favor of electronic records maintained by HUD for the purpose of verification of both the ownership and the insured status of a mortgage. As a result, HUD made several procedural changes that affected the originating lender, the holding lender, and the servicing lender.19 HUD determined that it was imperative that the data contained in HUD’s Single Family Insurance System regarding a lender’s FHA-insured portfolio be accurate.20 Of key concern was the submission of mortgage record changes and mortgage insurance terminations that update HUD’s insurance system. Lenders must notify HUD of a sale of an FHA-insured loan within 15 calendar days.21 19 Mortgagee Letter 2003-17 20 Mortgagee Letters 2003-17, 2004-34, 2005-11, and 2005-42 21 24 CFR 203.431, Sale of insured mortgage to approved mortgagee 15 HUD identified that the most common problem was that lenders often did not update the holder of record for each loan as required. As of December 1, 2005, only the existing holder of record is able to provide HUD with mortgage record changes to update a new holder of record if 90 days have passed after endorsement.22 Semper Took Immediate Corrective Action Semper officials acknowledged that they had not notified HUD or updated mortgage records upon the sale of FHA-insured loans because they were not aware of the requirements. However, they took immediate action on this finding. HUD will have to verify the updated mortgage records after the next refresh of data in HUD’s single-family systems. Conclusion Semper officials did not properly notify HUD upon the sale and/or transfer of FHA-insured loans. This condition occurred because the officials were not aware of the HUD requirements regarding mortgage record changes. Inaccurate or untimely reporting of mortgage record changes directly affects the payment of claims for insurance benefits. HUD will not pay a claim for insurance benefits for which the information on the claim and HUD’s FHA insurance system does not agree. Therefore, it is incumbent upon the lender to ensure that HUD’s records accurately reflect both the correct holder and servicer of record. Recommendations We recommend that HUD’s Deputy Assistant Secretary for Single Family Housing require Semper to 3A. Update its remaining mortgage records in HUD’s system to reflect the appropriate mortgage holder. 3B. Implement procedures to ensure the timely submission of mortgage record changes for future loans sold to investing lenders. 22 Mortgagee Letter 2005-42 16 SCOPE AND METHODOLOGY We identified Semper as a lender for review based on a risk assessment of mortgage lenders in the New England region. We researched lenders using HUD’s Single Family Neighborhood Watch system (Neighborhood Watch) and Single Family Housing Enterprise Data Warehouse system (Enterprise Data Warehouse). Neighborhood Watch is a Web-based comprehensive data processing, automated querying, reporting, and analysis system designed to highlight exceptions to lending practices to high-risk lenders and mortgages, so that potential problems that may arise are readily identifiable. Enterprise Data Warehouse is a data warehouse that is the key source of single-family data. It allows queries and provides reporting tools to support oversight activities, market and economic assessment, public and stakeholder communication, planning and performance evaluation, policies and guidelines promulgation, monitoring, and enforcement. Our audit period was January 1, 2008, through December 31, 2009. We identified lenders that Were active direct endorsement lenders, Had a home or branch office in Region 1 (New England states), Had originated at least 100 loans in the past 2 years, Had a higher percentage of loans that defaulted within the first 2 years after endorsement compared to the rest of the area selected for comparison, and Had not been reviewed by HUD’s Office of Inspector General (OIG) or HUD’s Quality Assurance Division in the past 5 years. To accomplish the survey objectives, we Identified, obtained, and reviewed relevant regulations pertaining to the origination of single-family mortgages, including the Code of Federal Regulations, HUD handbooks, mortgagee letters, and the United States Code. Obtained and reviewed pertinent performance information relating to the lender. Obtained, reviewed, and documented whether the lender had maintained a quality control plan that met the requirements of HUD Handbook 4060.1, REV-2. Obtained and reviewed copies of policies and procedures that the lender used for its loan origination processes. Reviewed HUD post endorsement technical review data. Identified and conducted a review of 22 FHA-insured loans (17 loans identified as seriously delinquent in HUD’s systems out of 38 that were early payment defaults plus and an additional 5 loans that were originated and fully underwritten by Semper as a full direct endorsement lender). Assessed other general aspects of the branch’s operations to ensure its continued lender approval status. We identified and conducted a detailed review of 17 FHA-insured loans, with a combined mortgage value of $3,839,415, originated by Semper. We selected the loans based on several risk factors from the 38 loans that went into early payment default within the first 2 years of origination during our audit period: 17 Loans that were claim terminated, Purchase loan transactions, Loans that went to claim with six or fewer payments before the first default being reported, Loans with excessive debt ratios, and Loans with gift letter amounts. The 17 loans represented the best loans for selection out of the 38 loans that were early payment defaults based on the analysis of available loan-level data and online records searches. Additionally, on April 14, 2010, Semper became a full underwriting lender (previously, the lender was a loan correspondent), and we selected an additional 5 FHA-insured loans, with a combined mortgage value of $1,196,221, for review from the 35 loans originated and underwritten by Semper through the July 31, 2010 reporting period, that had excessive debt ratios (greater than 31% front ration and 43% back ratio). This methodology allowed us to focus on loans that had a greater inherent risk to the FHA insurance fund and/or of noncompliance or abuse. We relied on information from systems used by HUD (including Neighborhood Watch and Enterprise Data Warehouse) to target loans for review and verified that the information submitted to HUD was consistent with the information in the lender’s own files. We also selected and reviewed 33 FHA-insure loans that closed between April 2009 and May 2010 that were reviewed under Semper’s quality control plan and determined the timeliness of the reviews, the adequacy of the reviews and whether they met HUD requirements. We conducted the audit in accordance with generally accepted government auditing standards. Those standards require that we plan and perform the audit to obtain sufficient, appropriate evidence to provide a reasonable basis for our findings and conclusions based on our audit objectives. We believe that the evidence obtained provides a reasonable basis for our findings and conclusions based on our audit objectives. 18 INTERNAL CONTROLS Internal control is a process adopted by those charged with governance and management, designed to provide reasonable assurance about the achievement of the organization’s mission, goals, and objectives with regard to Effectiveness and efficiency of operations, Reliability of financial reporting, and Compliance with applicable laws and regulations. Internal controls comprise the plans, policies, methods, and procedures used to meet the organization’s mission, goals, and objectives. Internal controls include the processes and procedures for planning, organizing, directing, and controlling program operations as well as the systems for measuring, reporting, and monitoring program performance. Relevant Internal Controls We determined that the following internal controls were relevant to our audit objectives: Loan origination process – Policies and procedures established by management to ensure that FHA-insured loans are originated in accordance with HUD requirements. Quality control process – Policies and procedures established by management to ensure that the quality control plan has been implemented and related reviews are performed in accordance with HUD requirements. We assessed the relevant controls identified above. A deficiency in internal control exists when the design or operation of a control does not allow management or employees, in the normal course of performing their assigned functions, the reasonable opportunity to prevent, detect, or correct (1) impairments to effectiveness or efficiency of operations, (2) misstatements in financial or performance information, or (3) violations of laws and regulations on a timely basis. 19 Significant Deficiencies Based on our review, we believe that the following items are significant deficiencies: Sponsors of Semper did not follow HUD requirements when underwriting two FHA-insured loans (see finding 1). Semper did not ensure that it adequately implemented its quality control plan (see finding 2). Semper did not ensure that its mortgage records were accurate in HUD’s systems (see finding 3). 20 APPENDIXES Appendix A SCHEDULE OF QUESTIONED COSTS AND FUNDS TO BE PUT TO BETTER USE Recommendation Ineligible 1/ Funds to be put number to better use 2/ 1A $179,400 1B $169,000 1/ Ineligible costs are costs charged to a HUD-financed or HUD-insured program or activity that are not allowable by law; contract; or Federal, State, or local policies or regulations. 2/ Recommendations that funds be put to better use are estimates of amounts that could be used more efficiently if an OIG recommendation is implemented. These amounts include reductions in outlays, deobligation of funds, withdrawal of interest, costs not incurred by implementing recommended improvements, avoidance of unnecessary expenditures noted in preaward reviews, and any other savings that are specifically identified. Implementation of our recommendation to require the sponsor of Semper to indemnify HUD for the loan that was not originated in accordance with HUD/FHA requirements will reduce FHA’s risk of loss to the insurance fund. The amount reflects that, upon the sale of the mortgaged property, FHA’s average loss experience is about 60 percent of the unpaid principal balance (see footnote 12). 21 Appendix B AUDITEE COMMENTS AND OIG’S EVALUATION Ref to OIG Evaluation Auditee Comments February 21, 2011 Mr. John A. Dvorak Regional Inspector General for Audit 10 Causeway St, Room 370 Boston MA, 02222 Dr. Mr. Dvorak; We have just recently had our exit interview and submit this document as our response to the findings. On a personal note I would like to say that all of the inspectors acted in a professional manner at all times and strived to be as unobtrusive as possible during the time they were on site at my office. That being said, overall I am pleased with the result of what is our first HUD audit. Although it wasn’t perfect, I do think it shows that our systems and oversight work and will continue to produce qualified and performing loans going forward. Finding 1: I will not comment on these issues since they relate to our Sponsors and not Semper. I will however use it as a cautionary tale. Finding 2: Our quality control plan was deficient. The main issue was timeliness. Our third party auditing firm, BSI Financial did not Comment 1 complete the audits within the specified time frame. Part of the reason was that the original firm we contracted with, Fraudmit went through some difficulty and referred us to another firm. I used some of the time between firms to find another company that may offer better services at a better price. Unfortunately I wasn’t able to commit enough time for that pursuit so I ended up going with BSI Financial, the firm that Fraudmit recommended. Fortunately the timing issue has been resolved, but despite that I have decided to use ComplianceEase beginning January 2011. 22 AUDITEE COMMENTS AND OIG’S EVALUATION Ref to OIG Evaluation Auditee Comments I believe they offer betters services and products at a slightly better price. There was also evidence that we didn’t follow all specifics detailed in our QC plan. This is an oversight on my part and I will be implementing increased training as well as further clarifications of written policies and procedures. I also agree with the recommendation that HUD follow up with us in 9 months to see how we are tracking. Finding 3: Mortgage Records not Accurate in HUD System. Again, we agree with this finding. At the time, we did not have a strict process for updating the FHA Connection Record Changes. Part of this was because no one person was tasked with the responsibility of making Comment 1 sure the record change took place in a timely manner. This has been resolved as the Sr. Processor is now responsible for updating the record changes in “real-time” as soon as the loan is transferred. In my opinion, as of right now, this violation has been corrected and the system tweaked to make sure that we are 100% compliant in this area from now on. In closing, the audit process was actually beneficial from our standpoint because it helped us identify potential areas of concern. We welcome any additional comments/suggestions you may have for us and will continue to work to improve our policies/procedures and training programs. Thank you; Michael J Securo EVP Compliance 23 OIG Evaluation of Auditee Comments Comment 1 The auditee acknowledged agreement with the findings and recommendations, and indicates corrective action which HUD will need to confirm implementation of and resolve the findings. 24 Appendix C HUD COMMENTS AND OIG’S EVALUATION Ref to OIG Evaluation HUD Comments On February 24, 2011, Denise Ramirez, a Management Analyst with HUD’s Quality Assurance Division, provided written comments. The comments were received via e-mail and are included, as written, below. “Good afternoon, John, Housing is concerned with the language used in Finding 1 which states, “A Sponsor of Semper Did Not Underwrite Two Loans in Accordance with HUD Requirements.” Whereas the loans cited in the audit report were originated by Semper Home Loans, the violations identified are the result of the underwriting decisions made by Fairfield Financial Mortgage Corporation. Comment 1 Therefore, it would not be appropriate to cite these violations and a recommendation for indemnifications in the Semper Home Loans audit report. It could be misleading and/or misrepresentative of Semper Home Loans’ role in the origination of the loans. Corrective action should be addressed directly to Fairfield Financial Mortgage Corporation, the sponsor for the referenced cases. Findings 2 and 3 will be addressed when the final audit is issued. I am available at (202) 402-8251 if you would like to discuss this further. Denise Ramirez Management Analyst HHQ Quality Assurance Division” 25 OIG Evaluation of Auditee Comments Comment 1 It is appropriate to identify or cite the violations of Semper's sponsor, Fairfield Financial Mortgage Group, Inc. (Fairfield Financial), in this report as it was a significant reportable finding that was identified during this audit. Also, the language is clear and is not misleading and/or misrepresentative of Semper Home Loans' role in the origination of loans, as the report makes it very clear that the violations cited were attributable to a sponsor of Semper and not Semper itself. Further, the recommendation is for HUD to seek from the sponsor of Semper for indemnification and reimbursement. However, to further distinguish Semper from its sponsor, Fairfield Financial, additional language was added to the body of the report, as well as the recommendation, referring to Fairfield Financial specifically. 26 Appendix D LOAN DETAILS FHA case no. Mortgage Unpaid HUD loss on Computed amount principal loan benefit of balance indemnification 451-0946313 $308,560 $299,000 $179,40023 451-0941164 $315,056 $309,686 $169,00024 Totals $623,616 $608,686 $169,000 $179,400 23 Data obtained on January 26, 2011, from HUD’s Neighborhood Watch/Early Warning System. HUD’s estimated loss is computed using FHA’s FY 2009 Actuarial Review of the Mutual Mortgage Insurance Fund. The average loss experienced is about 60% of the unpaid principal balance upon the sale of a mortgaged property. 24 Data obtained on January 26, 2011, from HUD’s Neighborhood Watch/Early warning System. HUD’s loss in this case is based on the actual loss incurred by HUD after foreclosure sale of the mortgaged property. 27
Semper Home Loans, Inc., Providence, RI, Needs To Improve Its Quality Control Process for Loan Origination and Updating of Mortgage Records
Published by the Department of Housing and Urban Development, Office of Inspector General on 2011-03-02.
Below is a raw (and likely hideous) rendition of the original report. (PDF)